|
Activity Number:
|
290
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #305663 |
|
Title:
|
Error Sum of Squares Comparison for Model Search, Identification, and Discrimination
|
|
Author(s):
|
Hongjie Deng*+ and Subir Ghosh
|
|
Companies:
|
University of California, Riverside and University of California, Riverside
|
|
Address:
|
3434 Kentucky Street, Riverside, CA, 92507,
|
|
Keywords:
|
class of models ; error sum of squares ; fractional factorial plans ; interaction effects ; linear model
|
|
Abstract:
|
We consider model search, identification, and discrimination between models within a class of models with common and uncommon parameters. We use the error sum of squares (SSEs) as our criterion function. We present general results for comparing SSEs of different models. We illustrate our method and results for comparing models in a factorial experiment involving m factors at two levels. The models considered have the common parameters as the general mean and the main effects and the uncommon parameters as two two-factor interactions, one in each model. Several fractional factorial plans are used.
|
